Pour tester les trois facteurs de Fama et French (FF) sur le marché des actions canadien, nous avons utilisé trois méthodes: l'analyse des statistiques descriptives, l'application de trois modèles unifactoriels relatifs aux trois facteurs (le portefeuille de marché, la Capitalisation Boursière (CB) et le ratio Valeur comptable/Valeur de marché (VC/VM)) et l'application du modèle à trois facteurs de FF. Notre étude prend la forme d'une analyse en séries temporelles des rendements de portefeuilles d'actions classés selon la CB et le ratio VC/VM. Notre contribution par rapport aux études canadiennes récentes (tel que celle de L'Her, Masmoudi et Suret (2002), Francoeur (2006) et Carmichael, Coënz et L'Her (2007)) consiste à classer et à construire les portefeuilles d'une manière adaptée mais différente. Nous nous différencions aussi par l'utilisation de bases de données plus récentes et beaucoup moins utilisées. Finalement, nous nous distinguons par l'utilisation des statistiques descriptives et de l'analyse univariée, surtout que ces deux méthodes ont été appliquées par FF et négligées par la majorité des études canadiennes récentes. À travers les trois méthodes utilisées, nous avons constaté que l'effet taille existe. Le ratio VC/VM a un effet sur le rendement sauf que nous n'avons pas observé le signe attendu pour la totalité des portefeuilles. Par ailleurs, l'effet taille domine celui du ratio VC/VM. D'un autre côté, le portefeuille du marché joue un rôle important dans l'explication du rendement. Et puis, nos résultats montrent qu'accompagné des deux autres déterminants, le rôle de la CB prend beaucoup plus d'ampleur. Par contre, avec le modèle multifactoriel, le portefeuille du marché a un rôle légèrement moins prononcé qu'avec le MÉDAF. D'un autre côté, même si la variabilité de rendement n'est pas totalement captée par les trois facteurs de FF, leur modèle décrit beaucoup mieux le rendement que les trois autres modèles unifactoriels. De plus, chacun des facteurs a un rôle dans la capture de la variabilité du rendement. En d'autres termes, il n'y a aucune variable de trop qui devrait être éliminée du modèle. ______________________________________________________________________________ MOTS-CLÉS DE L’AUTEUR : Modèle à trois facteurs de FF, Modèle d'évaluation des actifs financiers, Portefeuille du marché, Ratio valeur comptable/valeur de marché, Capitalisation boursière.

This thesis examines the (geometric) Steiner tree problem: Given a set of points P in the plane, find a shortest tree interconnecting all points in P, with the possibility of adding points outside P, called the Steiner points, as additional vertices of the tree. The Steiner tree problem has been studied in different metric spaces. In this thesis, we study the problem in Euclidean and rectilinear metrics. One of the most natural heuristics for the Steiner tree problem is to use a minimum spanning tree, which can be found in O(nlogn) time . The performance ratio of this heuristic is given by the Steiner ratio, which is defined as the minimum possible ratio between the lengths of a minimum Steiner tree and a minimum spanning tree. We survey the background literature on the Steiner ratio and study the generalization of the Steiner ratio to the case of obstacles. We introduce the concept of an anchored Steiner tree: an obstacle-avoiding Steiner tree in which the Steiner points are only allowed at obstacle corners. We define the obstacle-avoiding Steiner ratio as the ratio of the length of an obstacle-avoiding minimum Steiner tree to that of an anchored obstacle-avoiding minimum Steiner tree. We prove that, for the rectilinear metric, the obstacle-avoiding Steiner ratio is equal to the traditional (obstacle-free) Steiner ratio. We conjecture that this is also the case for the Euclidean metric and we prove this conjecture for three points and any number of obstacles.

Introduction: It is recommended to place all the vacuum tubes directly on a sample cradle after vein puncture to prevent analytic error. This recommendation is not always easy to follow because the samples are taken by different professionals under different situations.  The three most common analyses, platelets count, haemoglobin and prothrombin time were tested.  Therefore, it was interesting to compare results from the three most common analyses with or without sample cradle, to evaluate the influence of this step on the result. Methods: Three analyses were preformed, using blood from 50 different persons. Each person gave two vacuum tubes, each contained 4.5mL of venous blood for the study. Tubes containing EDTA were used for platelet counts and measurement of haemoglobin and tubes containing citrate were used for prothrombin time-analysis. One of the tubes was placed, as recommended, directly on the sample cradle while the other tube was placed flat on a bench for 10 minutes before it was placed on the sample cradle. Results: There was a clear difference in platelet counts with and without immediate cradling but only minor difference between the results for haemoglobin and International Normalized Ratio. Conclusion: Some analyses seem to be more sensitive for variation in cradling than others. For platelet count it was important to immediately rock the tubes but for determination of prothrombine time and hemoglobin it had a small impact. The small impact on the results is probably due to the efficiency of the anticoagulant in the vacuum tubes.

This thesis is concerned with caching algorithms. We investigate three variations of the caching problem: web caching in the Torng framework, relative competitiveness and caching with request reordering. <br /><br /> In the first variation we define different cost models involving page sizes and page costs. We also present the Torng cost framework introduced by Torng in [29]. Next we analyze the competitive ratio of online deterministic marking algorithms in the BIT cost model combined with the Torng framework. We show that given some specific restrictions on the set of possible request sequences, any marking algorithm is 2-competitive. <br /><br /> The second variation consists in using the relative competitiveness ratio on an access graph as a complexity measure. We use the concept of access graphs introduced by Borodin [11] to define our own concept of relative competitive ratio. We demonstrate results regarding the relative competitiveness of two cache eviction policies in both the basic and the Torng framework combined with the CLASSICAL cost model. <br /><br /> The third variation is caching with request reordering. Two reordering models are defined. We prove some important results about the value of a move and number of orderings, then demonstrate results about the approximation factor and competitive ratio of offline and online reordering schemes, respectively.

This thesis examines the (geometric) Steiner tree problem: Given a set of points P in the plane, find a shortest tree interconnecting all points in P, with the possibility of adding points outside P, called the Steiner points, as additional vertices of the tree. The Steiner tree problem has been studied in different metric spaces. In this thesis, we study the problem in Euclidean and rectilinear metrics. One of the most natural heuristics for the Steiner tree problem is to use a minimum spanning tree, which can be found in O(nlogn) time . The performance ratio of this heuristic is given by the Steiner ratio, which is defined as the minimum possible ratio between the lengths of a minimum Steiner tree and a minimum spanning tree. We survey the background literature on the Steiner ratio and study the generalization of the Steiner ratio to the case of obstacles. We introduce the concept of an anchored Steiner tree: an obstacle-avoiding Steiner tree in which the Steiner points are only allowed at obstacle corners. We define the obstacle-avoiding Steiner ratio as the ratio of the length of an obstacle-avoiding minimum Steiner tree to that of an anchored obstacle-avoiding minimum Steiner tree. We prove that, for the rectilinear metric, the obstacle-avoiding Steiner ratio is equal to the traditional (obstacle-free) Steiner ratio. We conjecture that this is also the case for the Euclidean metric and we prove this conjecture for three points and any number of obstacles.

Digital image and video in their raw form require an enormous amount of storage capacity. Considering the important role played by digital imaging and video, it is necessary to develop a system that produces high degree of compression while preserving critical image/video information. There are various transformation techniques used for data compression. Discrete Cosine Transform (DCT) and Discrete Wavelet Transform (DWT) are the most commonly used transformation. DCT has high energy compaction property and requires less computational resources. On the other hand, DWT is multiresolution transformation.<p> In this work, we propose a hybrid DWT-DCT algorithm for image compression and reconstruction taking benefit from the advantages of both algorithms. The algorithm performs the Discrete Cosine Transform (DCT) on the Discrete Wavelet Transform (DWT) coefficients. Simulations have been conducted on several natural, benchmark, medical and endoscopic images. Several QCIF, high definition, and endoscopic videos have also been used to demonstrate the advantage of the proposed scheme.<p> The simulation results show that the proposed hybrid DWT-DCT algorithm performs much better than the standalone JPEG-based DCT, DWT, and WHT algorithms in terms of peak signal to noise ratio (PSNR), as well as visual perception at higher compression ratio. The new scheme reduces false contouring and blocking artifacts significantly. The rate distortion analysis shows that for a fixed level of distortion, the number of bits required to transmit the hybrid coefficients would be less than those required for other schemes Furthermore, the proposed algorithm is also compared with the some existing hybrid algorithms. The comparison results show that, the proposed hybrid algorithm has better performance and reconstruction quality. The proposed scheme is intended to be used as the image/video compressor engine in imaging and video applications.

The Swedish Association of Local Authorities and Regions (SALAR) and the Swedish National Board of Health and Welfare (NBHW) have published a yearly report called “Open Comparisons of Quality and Efficiency in Swedish Health Care –Regional Comparisons” since the year 2006 to compare medical outcomes, patient experience, availability and costs among hospitals and counties. In this paper, inspired by the region profiles attached to the report, we describe several possible measurement methods for comparing dichotomous data: Risk Difference (RD), Relative Risk (RR), Relative Risk Reduction (RRR) and Odds Ratio (OR), mainly by using a geometric approach as a basis for further discussion. A standard scores method is also presented as a means for more efficient comparisons.

The human epidermal growth factor receptor-2 (HER2) protein has been associated with breast cancer progression and the HER2 status can be used to determine the type of treatment for each breast cancer patient. The purpose of this study was to examine the HER2 protein and gene statuses in breast cancer tissue samples using two methods and analyze the concordance between them. Ten paraffin-embedded, formaldehyde-fixed breast cancer tissue samples from the Biobank at the Department of Pathology and Cytology at Sundsvall Hospital were analyzed in this study. All samples were from women born between 1931 and 1976. The methods used were immunohistochemistry (IHC) to visualise the HER2 protein and silver in situ hybridization (SISH) to detect gene amplification. The IHC staining method is an indirect detection of the HER2 protein using antibodies. The SISH method used in this study is a Dual ISH which detects both the HER2 gene and the centromere region of Chromosome 17 on the same tissue slide. A HER2 gene/Chromosome 17 ratio was calculated according to the manufacturer’s instructions. This ratio was used to determine HER2 gene status. Out of ten samples, seven were positive with IHC and three were negative. The results from the SISH staining exposed a gene amplification in three of the IHC positive samples, while seven samples did not contain any amplified HER2 genes. The conclusion was that the concordance between IHC and SISH for HER2 was 60 percent.

In chapter 1, we show (i) that the risk-return characteristics of our sample of 17 developed stock markets of the world have converged significantly toward each other during our study period 1974 2004, and (ii) that this international convergence in risk-return characteristics is driven mainly by the declining country effect, rather than the rising industry effect, suggesting that the convergence is associated with international market integration. Specifically, we first compute the risk-return distance among international stock markets based on the Euclidean distance and find that the distance thus computed has been deceasing significantly over time, implying a mean-variance convergence. In particular, the average risk-return distance has decreased by about 43% over our sample period. The speed of convergence, however, varies greatly across individual markets, largely reflecting the initial distance of each individual market from the international average risk-return characteristic. Lastly, we document that the risk-return characteristics of our sample of 14 emerging markets have been converging rapidly toward those of developed markets in recent years. This development notwithstanding, emerging markets still remain as a distinct asset class. In chapter 2, we examine the historical evolution of international earnings-to-price ratios for a sample of 17 markets over the period 1980 2004. We introduce a distance measure of earnings-to-price ratios among international stock markets and find that earnings-to-price ratios of 17 markets have significantly converged toward each other during the period. The average distance measure for 17 markets has decreased by about 80 percent during the period. The speed of convergence for individual markets varies and mainly reflects the initial distance of individual markets from the international average. We also find that although both country and industry effects account for convergence in earnings-to-price ratios among the sample markets, country effect dominates industry effect in terms of the magnitude. We further examine what could explain the declining country effect and document that the time trend of dividend-yield distance measure closely follows that of earnings-to-price distance measure. This result suggests that convergence in earnings-to-price ratio is mainly due to convergence in economic factors such as growth opportunities or discount rates rather than due to convergence in accounting practices.

Numerical study of laminar convective cooling of ribs in a parallel plate channel is investigated. Single rib mounted on one channel wall in forced, mixed and free convection is analyzed. Furthermore, the series ribs array with in-line and staggered mounted on channel walls are considered. Through the use of a stream function vorticity transformation, solution of the transformed governing equations for the system is obtained using the control volume method with non-uniform grid. The effects of the Reynolds number, thermal conductivity ratio of rib to fluid and rib¡¦s profile area on heat transfer rate of single rib and rib array are presented. In addition, the effects of the length from inlet to the first rib and the space between ribs for rib array are carried out. A correlation for single and rib array in forced convection is presented to estimate the optimum aspect ratio of rib with various Reynolds number, thermal conductivity ratio of rib to fluid, rib¡¦s profile area. Furthermore, the results of different Gr/Re2 and various channel inclination angle in mixed convection are also examined numerically. The results indicate that both in forced and mixed convection, the optimum aspect ratio of a rib corresponding to the rib with maximum heat transfer rate increases with increasing Re but decreases with K for a fixed rib profile area. In forced convection the optimum aspect ratio of rib array increases with rib¡¦s space but decreases with the length from inlet to the first rib of channel. Then, numerical correlations to predict the optimum aspect ratio of single rib and rib array are developed for fixed rib¡¦s area with various Re, K and rib number. In mixed convection, the optimum aspect ratios of single rib and staggered rib array increase with not only the inclination angle but also Gr/Re2.